¹ Bourdieu, P., Language and Symbolic Power II 4, ed. Thompson, J.B., transl. Raymond, G. and Adamson, M. (Cambridge, 1991), 120Google Scholar; original version ‘Les rites comme actes d'institution’, Actes de la recherche en sciences sociales 43 (Rites et fétiches) (1982), 58–63, at 60. I am grateful to the CQ reader for helpful comments and corrections.

² Those citizens who have a ‘liberal education’ (ἐλευθέριος παιδεία, Arist. Pol. 1338a30) that is not directed at making a living. See Raaflaub, K., ‘Democracy, oligarchy, and the concept of the “free citizen” in late fifth-century Athens’, Political Theory 11 (1983), 517–44CrossRefGoogle Scholar.

³ Regional koinai developed in north-west Greece (the Aetolian league), the north-central Peloponnese (the Achaean League), Sicily and Rhodes. See Bubeník, V., ‘Formation of Doric koines’, in Giannakis, G.K. et al. (edd.), Encyclopedia of Ancient Greek Language and Linguistics (Leiden, 2013), 1.603–6Google Scholar; Striano, A., ‘Koiné, koiná, koinaí: are we talking about the same thing?’, in Giannakis, G.K. et al. (edd.), Studies in Ancient Greek Dialects (Berlin, 2017), 131–47Google Scholar.

⁴ Kretschmer, P., ‘Zur Geschichte der griechischen Dialekte’, Glotta 1 (1909), 1–59Google Scholar.

⁵ Parker, H.N., ‘The linguistic case for the Aiolian migration reconsidered’, Hesperia 77 (2008), 431–64CrossRefGoogle Scholar, with a response by Ramón, J.L. García, ‘On the genetic classification of ancient Greek dialects: comparative reconstruction versus hypercriticism and atomism at work’, Studies in Greek Linguistics 30 (2010), 219–36Google Scholar; MacSweeney, N., ‘Separating fact from fiction in the Ionian migration’, Hesperia 86 (2017), 379–421Google Scholar; Pulgram, E., ‘Linear B, Greek and the Greeks’, Glotta 38 (1960), 171–81Google Scholar had already set out the contradictions inherent in the extension of nomenclature (names of peoples and languages) to undocumented periods in the construction of aetiologies.

⁶ As Étienne Bonnot de Condillac wrote: ‘Tout confirme donc que chaque langue exprime le caractère du peuple qui la parle’ (Essai sur l'origine des connaissances humaines [Paris, 1798]; Seconde partie, du langage et de la méthode. Section première, ch. 15 ‘Du génie des langues’, at 143). This echoes an idea about language that can be seen in Isocrates, and was ascribed to Ennius (Gell. NA 17.17). For Condillac it was part of the revolutionary programme to teach the standard language to the masses, since without a proper language they would never be able to achieve their full potential; but a side effect of a standard language, as Bourdieu observed, is the advantage acquired by the social class for whom it is a first language (at 47 n. 1; original version Ce que parler veut dire [Paris, 1982], 30).

⁷ Bourdieu (n. 1), 93; original version ‘Vous avez dit “populaire”?’, Actes de la recherche en sciences sociales 46 (L'usage de la parole) (1983), 98–105, at 100.

⁸ Meillet, A., Aperçu d'une histoire de la langue grecque (Paris, 1913), 104–6Google Scholar. The north-western dialects are often included in the term Doric by modern linguists, though speakers (e.g. in Elis, Phocis) did not necessarily regard themselves as Dorians. Dosuna, J. Méndez (Los dialectos dorios del noroeste [Salamanca, 1985], 316–26Google Scholar) has a useful review of the evidence and makes the case for the Doric future in -se- as an innovation exclusive to ‘protodorio’.

⁹ Brixhe, C., ‘Situation, spécificités et contraintes de la dialectologie grecque’, in id. and Vottéro, G. (edd.), Peuplements et genèses dialectales dans la Grèce antique (Nancy, 2006), 39–69Google Scholar. See Ramón, J.L. García, ‘Ancient Greek dialectology: old and new questions, recent developments’, in Giannakis, G.K. et al. (edd.), Studies in Ancient Greek Dialects (Berlin, 2017), 29–106Google Scholar for a defence of the traditional groupings.

¹⁰ ‘Nous verrons … que dans sa volonté de mettre de l'ordre dans le désordre la linguistique a parfois tendance à dériver vers le rêve d'un ordre supérieur, d'une langue originelle, voire d'un «intelligent design». Un regard sur l'histoire récente, celle pour laquelle nous avons des traces tangibles, nous donne pourtant une autre leçon.’ Calvet, L.-J., ‘Pour une linguistique du désordre et de la complexité’, in Blanchet, P., Calvet, L.-J., de Robillard, D. (edd.), Un siècle après le Cours de Saussure, la linguistique en question (Paris, 2007), 13–80, at 25Google Scholar.

¹¹ These models can be seen in ancient Greek sources from Herodotus to Strabo, which regularly explain language change (and the emergence of a new ethnic consciousness) by reference to the movement of a large population into a region (e.g. Hdt. 1.56–7, Thuc. 1.12, Strabo 8.1.2). There is nuance in the ancient sources, however: they also invoke language contact as a cause of change, and Herodotus in discussing the language of the Sauromatae (4.114–17), which is a peculiar type of Scythian, explains that the Amazons learned it imperfectly (and by implication passed on a modified version to their children). Imperfect learning is advanced as a cause of linguistic change in contexts of language shift by Thomason, S.G. and Kaufman, T., Language Contact, Creolization, and Genetic Linguistics (Berkeley, 1988), 38CrossRefGoogle Scholar and by others.

¹² The departure of Columbus for the New World was framed by the first Latin edition of Herodotus in 1474 (Venice, Jacobus Rubeus) and the first Greek edition in 1502 (Venice, Aldus).

¹³ Anatolian is most often suggested: see Finkelberg, M., Greeks and Pre-Greeks. Aegean Prehistory and Greek Heroic Tradition (Cambridge, 2005), 42–54Google Scholar for arguments.

¹⁴ See Thomason, S.G., Language Contact: An Introduction (Edinburgh, 2001), 129Google Scholar for the distinction between borrowing and shift-induced interference, a distinction which ‘correlates robustly with linguistic effects: non-basic vocabulary first and most in borrowing, with structure and basic vocabulary borrowed later if at all; phonology and syntax most prominent in shift-induced interference, with lexical transfer lagging behind or absent altogether. One implication of this distinction is that it permits an educated guess about the type of contact that was responsible for contact-induced changes in a long-vanished contact situation’.

¹⁵ Del Valle, J., ‘Language, politics and history’, in id. (ed.), A Political History of Spanish (Cambridge, 2013), 3–20, at 18CrossRefGoogle Scholar.

¹⁶ Pulgram, E., ‘The nature and use of proto-languages’, Lingua 10 (1961), 18–37, at 29CrossRefGoogle Scholar: ‘Instead it seems highly probable and reasonable … not that Latin was imported in the Iberian peninsula and there existed for a while in a modified form called Proto-Ibero-Romanic before it was somehow decomposed into various dialects, but that Latin (and very likely not just one kind of local or social dialect of Latin, nor one Latin of a single period) was superimposed upon, and exposed to the substratic influence of, a variety of already existing dialects … That is to say, there never were any people to whom Proto-Ibero-Romanic was a native language.’ See also Pulgram, E., ‘Spoken and written Latin’, Language 26 (1950), 458–66CrossRefGoogle Scholar on ‘vulgar Latin’; and Wright, R., Late Latin and Early Romance in Spain and Carolingian France (Liverpool, 1982), ch. 3Google Scholar ‘Carolingian France: the invention of Medieval Latin’.

¹⁷ What H. Kloss, in a fundamental study of the notion of dialect and Dachsprache, called the ‘near-dialectization of a sister language’: ‘Abstand languages and Ausbau languages’, Anthropological Linguistics 9 (1967), 29–41Google Scholar.

¹⁸ ‘Outside linguistics, the term has … undergone a reinterpretation from a geographic characterisation to a qualitative ranking: in general usage, Hochdeutsch is commonly understood to refer to a “higher” form of language, a culturally elevated Hochsprache “High language” superior to other forms of German. This reinterpretation establishes a particularly powerful case of standard language ideology.’ Wiese, H., ‘“This migrants’ babble is not a German dialect!” The interaction of standard language ideology and “us” / “them” dichotomies in the public discourse on a multiethnolect’, Language in Society 44 (2015), 341–68, at 345–6CrossRefGoogle Scholar.

¹⁹ Hudson, R.A., Sociolinguistics (Cambridge, 1980), 34Google Scholar.

²⁰ J. and Milroy, L., ‘Mechanisms of change in urban dialects: the role of class, social network and gender’, International Journal of Applied Linguistics 3 (1993), 57–77CrossRefGoogle Scholar.

²¹ Papyri are a rich source of linguistic variation for Egyptian Greek from the Hellenistic period until the Arab invasions; in addition to diachronic development, variation can often be correlated with ethnicity, occupation and socio-economic status. See Depauw, M. and Stolk, J., ‘Linguistic variation in Greek papyri: towards a new tool for quantitative study’, GRBS 55 (2015), 196–220Google Scholar; Evans, T. and Obbink, D. (edd.), The Language of the Papyri (Oxford, 2010)Google Scholar.

²² Osborne, R., Demos: The Discovery of Classical Attika (Cambridge, 1985), 42Google Scholar; Hansen, M.H., The Athenian Democracy in the Age of Demosthenes (Oxford, 1991), 93Google Scholar; with Hansen, M.H., Studies in the Population of Aigina, Athens and Eretria (Copenhagen, 2006)Google Scholar.

²³ I take the term thēs (with hippeis ‘cavalry’ and zeugitai ‘hoplite class’) to reflect not a Solonic property assessment but an index of current economic capability. See de Ste. Croix, G.E.M., ‘The Solonian census classes and the qualifications for cavalry and hoplite service’, in Harvey, D. and Parker, R. (edd.), Athenian Democratic Origins and Other Essays by G.E.M. de Ste. Croix (Oxford, 2004), 46–56CrossRefGoogle Scholar; van Wees, H., ‘The myth of the middle-class army: military and social status in ancient Athens’, in Bekker-Nielsen, T. and Hannestad, L. (edd.), War as a Cultural and Social Force: Essays on Warfare in Antiquity (Copenhagen, 2001), 45–71Google Scholar.

²⁴ E.g. the innovative dative plural -αις of the first declension was suppressed until the 420s: see Dover, K.J., ‘The language of classical Attic documentary inscriptions’, TPhS 79 (1981), 1–14Google Scholar, repr. in Dover, K.J., Greek and the Greeks (Collected Papers, vol. 1 [Oxford, 1987]), 31–41Google Scholar.

²⁵ For the convenience of readers, and in the interests of balance, references are given to appropriate sections of both Threatte, L., The Grammar of the Attic Inscriptions, vol. 1: Phonology (Berlin, 1980)Google Scholar and Teodorsson, S.-T., The Phonemic System of the Attic Dialect 400–340 b.c. (Göteborg, 1974)Google Scholar. Threatte's grammar does not, on the whole, concern itself with ‘substandard’ Attic, and he is suspicious of the early dating of sound changes which are traditionally associated with the koinē.

²⁶ When the old diphthongs [ei, ou] became long close vowels [e:, o:], they merged with existing long close vowels [e:, o:] which were the result of contraction and compensatory lengthening. As a result, the spellings ΕΙ and ΟΥ came to be used for all of these vowels: thus φιλεῖτε from φιλέ-ετε. The term ‘spurious diphthong’ has traditionally been used by classicists for this digraph spelling. Synchronically, of course, the Athenians would have had no idea whether the sound written ΕΙ or ΟΥ had been a real diphthong or not. Close [e] was approximately the vowels in French été, close [o] as in French mot.

²⁷ Immerwahr's corpus numbers 934, 1075, 2620, 4472, 5592, 6149, 6720, 8100 (a couple of these may be miswritten rather than significant): H. Immerwahr, Corpus of Attic Vase Inscriptions (PDF available online: version dated January 2009).

²⁸ Colvin, S., ‘The language of non-Athenians in Old Comedy’, in Harvey, D. and Wilkins, J. (edd.), The Rivals of Aristophanes (London, 2000), 285–98, at 288–91Google Scholar; Cassio, A.C., ‘Attico “volgare” e Ioni in Atene alla fine del 5 secolo ac’, Annali Ist. Orient. Napoli 3 (1981), 79–93Google Scholar.

²⁹ Thus κύρρος < κύριος, etc. W. Blümel, Die aiolischen Dialekte (Göttingen, 1982), 55.

³⁰ See Dosuna, J. Méndez, ‘El cambio de <ε> en <ι> ante vocal en los dialectos griegos’, in Crespo, E. et al. (edd.), Dialectologica Graeca: Actas del II Coloquio Internacional de Dialectología Griega (Madrid, 1993), 237–59Google Scholar with additional bibliography.

³¹ Brenne, S., ‘Ostraka and the process of ostrakophoria’, in Coulson, W. (ed.), The Archaeology of Athens and Attica under the Democracy (Oxford, 1994), 13–24, at 21Google Scholar; Brenne, S., ‘Die Ostraka (487–ca. 416 v. Chr.) als Testimonien’, in Siewert, P. (ed.), Ostrakismos–Testimonien, vol. 1 (Stuttgart, 2002), 36–166, at 97 no. T 1/79Google Scholar. Discussed in Colvin, S., ‘Social dialect in Attica’, in Penney, J.H.W. (ed.), Indo-European Perspectives. Studies in Honour of Anna Morpurgo Davies (Oxford, 2004), 95–108Google Scholar.

³² Lang, M., Graffiti and Dipinti (Princeton, 1976)CrossRefGoogle Scholar.

³³ Lang, M., Ostraka (Princeton, 1990), 14–15Google Scholar.

³⁴ As Greek linguistic culture moved from koinē to Atticism in the Imperial period, elements of vernacular Greek were regularly stigmatized as barbarismos or soloikismos (from Soloi in Cilicia).

³⁵ Brixhe has shown that many of the ‘barbarisms’ uttered by foreigners in Aristophanes were also (innovative) features of non-standard Greek: Brixhe, C., ‘La langue de l’étranger non grec chez Aristophane’, in Lonis, R. (ed.), L’étranger dans le monde grec (Nancy, 1988), 113–38Google Scholar; ‘Les “ardoises” de l'Académie. Histoire exemplaire d'un dossier délicat’, in Dubois, L. and Masson, E. (edd.), Philokypros (Salamanca, 2000), 61–89, at 83–6Google Scholar.

³⁶ See Trudgill, P., Sociolinguistics: An Introduction to Language and Society (Penguin, 2001 ⁴), 61–80Google Scholar for an overview; and Dunn, M., ‘Gender determined dialect variation’, in Corbett, G.G. (ed.), The Expression of Gender (Berlin, 2014), 39–68Google Scholar.

³⁷ Willi, A., The Languages of Aristophanes (Oxford, 2003), 161–96Google Scholar; and see further below.

³⁸ Sommerstein, A.H., ‘The language of Athenian women’, in Sommerstein, A.H. and De Martino, F. (edd.), Lo spettacolo delle voci (Bari, 1995), 2.61–85Google Scholar; reprinted in Sommerstein, A.H., Talking about Laughter (Oxford, 2009), 1–14CrossRefGoogle Scholar.

³⁹ McClure, L.K., ‘Female speech and characterization in Euripides’, in Sommerstein, A.H. and De Martino, F. (edd.), Lo spettacolo delle voci (Bari, 1995), 2.35–59Google Scholar.

⁴⁰ Lakoff, R., ‘Language and woman's place’, Language in Society 2 (1973), 45–80, at 55–6CrossRefGoogle Scholar.

⁴¹ Duhoux, Y., ‘Langage de femmes et d'hommes en grec ancien: l'exemple de Lysistrata’, in Penney, J.H.W. (ed.), Indo-European Perspectives: Studies in Honour of Anna Morpurgo Davies (Oxford, 2004), 131–45, at 140Google Scholar. See also Willi (n. 37), 176–95 for analysis of comic data.

⁴² Classical scholars sometimes compare Cic. De or. 3.12.45: equidem cum audio socrum meam Laeliam—facilius enim mulieres incorruptam antiquitatem conseruant, quod multorum sermonis expertes ea tenent semper, quae prima didicerunt—, sed eam sic audio, ut Plautum mihi aut Naeuium uidear audire; sono ipso uocis ita recto et simplici est, ut nihil ostentationis aut imitationis afferre uideatur; ex quo sic locutum esse eius patrem iudico, sic maiores. ‘For my part, when I hear my mother-in-law Laelia (for it is easier for women to preserve uncorrupted [the language of] antiquity, because, by keeping less company than men, they always preserve what they first learned)—when I hear her I have the impression that I am hearing Plautus or Naevius: the sound of her voice is so correct and unaffected that she appears quite devoid of ostentation or affectation; from which I conclude that her father spoke in the same way, and her ancestors.’ This tells us about Roman ideas on the role of women as safeguards of purity, tradition and the Roman family, but not whether the Latin of aristocratic women in Rome was regularly marked by conservative linguistic features; especially as archaic Latin was less standardized than the classical language, and presents features that are paradoxically ‘innovative’ compared to classical Latin (where they are suppressed, but reappear in post-classical Latin or in informal inscriptions).

⁴³ Milroy (n. 20).

⁴⁴ ‘A literate woman must have been the exception and not the rule’ (S.G. Cole, ‘Could Greek women read and write?’ Women's Studies 8 [1981], 129–55, at 135). M.P.J. Dillon makes a case that aristocratic female literacy may have been more widespread than previously assumed, in ‘Engendering the scroll: girls’ and women's literacy in classical Greece’, in Grubbs, J. Evans and Parkin, T. (edd.), The Oxford Handbook of Childhood and Education in the Classical World (Oxford, 2013), 396–417CrossRefGoogle Scholar.

⁴⁵ So, for example, Vesp. 492–9 (a woman gives a feisty response to a customer asking for free onions), Ran. 549–78 (the innkeeper threatens to smash Xanthias’ teeth).

⁴⁶ Sommerstein (n. 38), 83.

⁴⁷ Willi (n. 37), 161–96.

⁴⁸ See S.-T. Teodorsson for a temporary reversal of the merging of certain front vowels in the fourth century: The Phonology of Attic in the Hellenistic Period (Göteborg, 1978), 68, 92–4Google Scholar.

⁴⁹ Isoc. Paneg. 50, Antid. 296.